Skip to content

Benchmarks

Sources
8
Entries tracked
22,310
Last sync
18 min
All sources syncing normally
  • Arena · Text6 hours ago
  • Arena · WebDev6 hours ago
  • Arena · Vision6 hours ago
  • Arena · Agent6 hours ago
  • GPQA Diamond19 minutes ago
  • SWE-bench18 minutes ago
  • Terminal-Bench18 minutes ago
  • ARC-AGI-218 minutes ago
Overall ×1WebDev Arena ×1Vision Arena ×1Factuality Arena ×1Search Arena ×1Document Arena ×1Agent Arena ×1GPQA Diamond ×1AIME (math) ×1FrontierMath ×1SimpleQA Verified ×1ARC-AGI-2 ×1Humanity's Last Exam ×1SimpleBench ×1LiveBench ×1BFCL v4 (function calling) ×1dormantSWE-bench Verified (live board) ×1Terminal-Bench 4.0 ×1MMLU-Pro (Artificial Analysis) ×1LiveCodeBench (Artificial Analysis) ×1IFBench (Artificial Analysis) ×1HHEM Hallucination (Vectara) ×19 of 1034 models
ModelOrgIntervalSwingiiiiiiiiiiiiiiiiiiiiiiAdd to compare
81grok-4.20-multi-agent-beta-0309xaiOct 2026—64.3 75.1Ranked 64.3 = (4 × 75.1 + 3 × 50) ÷ 7. 75.1 is the mean percentile over the 4 of 22 selected axes this model is measured on; fewer axes means a stronger pull toward 50.61–67234/22#48–8690—707467—————————————————
98Grok 4.20 0309 (Reasoning)xai——62.4 99.7Ranked 62.4 = (1 × 99.7 + 3 × 50) ÷ 4. 99.7 is the mean percentile over the 1 of 22 selected axes this model is measured on; fewer axes means a stronger pull toward 50.——1/22low—————————————————————100—
102Grok 4.20 0309 v2 (Reasoning)xai——62.2 98.8Ranked 62.2 = (1 × 98.8 + 3 × 50) ÷ 4. 98.8 is the mean percentile over the 1 of 22 selected axes this model is measured on; fewer axes means a stronger pull toward 50.——1/22low—————————————————————99—
119grok-4.20-beta1xaiOct 2026—61.4 72.8Ranked 61.4 = (3 × 72.8 + 3 × 50) ÷ 6. 72.8 is the mean percentile over the 3 of 22 selected axes this model is measured on; fewer axes means a stronger pull toward 50.58–64353/22#46–11992——6957—————————————————
207grok-4.20-beta-0309-reasoningxaiSep 2026—57.5 61.3Ranked 57.5 = (6 × 61.3 + 3 × 50) ÷ 9. 61.3 is the mean percentile over the 6 of 22 selected axes this model is measured on; fewer axes means a stronger pull toward 50.55–59666/22#49–20790327372—24————————77———————
259grok-4-20xAI—$1.25 / $2.5055.8 73.0Ranked 55.8 = (1 × 73.0 + 3 × 50) ÷ 4. 73.0 is the mean percentile over the 1 of 22 selected axes this model is measured on; fewer axes means a stronger pull toward 50.——1/22low#44–259———————————73——————————
287grok-4.20-0309-reasoningxAI——54.7 59.4Ranked 54.7 = (3 × 59.4 + 3 × 50) ÷ 6. 59.4 is the mean percentile over the 3 of 22 selected axes this model is measured on; fewer axes means a stronger pull toward 50.52–58543/22#68–287———————7975—24———————————
551Grok 4.20 Beta (Reasoning, 0309)xaiOct 2026—45.8 33.1Ranked 45.8 = (1 × 33.1 + 3 × 50) ÷ 4. 33.1 is the mean percentile over the 1 of 22 selected axes this model is measured on; fewer axes means a stronger pull toward 50.45–46—1/22low#96–551—33————————————————————
616grok-4.20-beta-0309-non-reasoningxAI——44.0 26.1Ranked 44.0 = (1 × 26.1 + 3 × 50) ÷ 4. 26.1 is the mean percentile over the 1 of 22 selected axes this model is measured on; fewer axes means a stronger pull toward 50.——1/22low———————————————26———————

Blend is coverage-shrunk; the raw composite is shown greyed beside it · A low badge means fewer than 3 boards scored this model

The few things in AI and quantum that actually mattered this week — honestly summarised.

One email a week. Unsubscribe in one click. No tracking, no sharing.